Captive portal/interface stops working under certain conditions
-
Hi all,
I'm using pfsense with captive portal since years now but over years/releases it still stops working randomly.
I have two sites with HA/CARP configuration each (pfsense on two different Proxmox servers):
SiteA -> Proxmox1-> PFSense1
SitaA -> Proxmox2-> PFSense2
SiteB -> Proxmox1-> PFSense1
SitaB -> Proxmox2-> PFSense2It happens that CaptivePortal is not working, no login page and no connectivity even for the allowed IPs that should work without authentication.
At that time, all the other interfaces/networks (up to 10) still work without issues.
A simple "save" of the captive portal config, so to say a restart, restores everything to normal.
It happens on both sites but not at the same time/day.
I tried to debug as per offical docs but, during the outage, the rules and the anchors are the same as pre/after restart.It was happening with 2.6.x release and it happens with 2.7.x so it seems not connected to the ipfw or pf way of blocking.
Reflecting on what happened during the previous hours/day, it seems somehow connected to the Carp switchover or interface status change.
It happens, even if not every time, when I restart to standby and active nodes for VM backups or when there's a CARP status change (lan disconnection, etc).Can anyone please suggest what could I look at to debug the outage next time?
thanks
-
@Neverstopdreaming too bad no one ever answered. I'm having the same problem, and it started after I configured carp HA. After entering CP settings and saving without doing any changes it starts working again, like your said. Logs after doing this show a check_reload_status activity followed by a minicron "(/etc/rc.prunecaptiveportal) terminated by signal 15 (Terminated)" message that is what actually gets it back to working.